Cerebrovascular Reactivity: Purpose, Optimizing Methods, and Limitations to Interpretation – A Personal 20-Year Odyssey of (Re)searching

Figures

• 

  Figure 1. Organization of cerebral blood flow (CBF) and collateral blood flow. CBF is supplied by four extracranial arteries originating from the aortic arch and subclavian arteries, the internal carotid arteries (ICA), and the vertebral arteries (VA). The latter joining to form the basilar artery (BA). Intracranially, these arteries anastomose around the circle of Willis, and separate into the major intracranial vessels perfusing each hemisphere: the anterior cerebral artery (ACA), middle cerebral artery (MCA), and posterior cerebral artery (PCA). Although there are multiple anastomoses between the major cerebral vascular territories, the blood supply to each territory, in baseline states, comes predominantly from its overlying artery (illustrated by color coding). In the case of partial or complete blockage of one of the cerebral arteries, the availability of blood flow depends on the ability to recruit collateral pathways. Figure modified from Fisher et al. (2018).

• 

  Figure 2. Pathophysiology of steal, and steal as a marker of availability of collateral flow. Left panel shows angiograms from two patients with nearly occluded carotid arteries (red arrows). Next to each are axial slice cerebrovascular reactivity (CVR) maps from the hypercapnic test. The CVR for each voxel is color coded according to the color scheme shown and mapped onto the corresponding voxel of the anatomical scan. The upper panel shows steal physiology (blue voxels on axial scan), and the lower panel shows near normal CVR. The suggested mechanism of each map is shown in the schematic to the right. Upper right panel: (1) Flow capacity of intracranial vessels greater than can be supplied by extracranial vessels depicted as “inflow limitation”; (2) occlusion of a branch; (3) hypercapnia and stimulation of vasodilation; (4) vasodilatory reserve (gray area) is exhausted downstream from the occlusive lesion (indicated by the red lines on the outside of the gray zone), but intact on the contralateral side (indicated by black lines inside gray zone); and (5) with hypercapnia, flow is directed predominantly to the contralateral side, dropping the perfusion pressure at the bifurcation. The…

• 

  Figure 3. The effect of partial pressure of carbon dioxide (PaCO2) on blood flow at different perfusion pressures. Data taken from Harper and Glass (1965). In this experiment, the authors anesthetized dogs and measured CBF by washout of radioactive counts. Graphs in the top row were taken from the paper and adjusted to the same scale. (A) Normotensive; (B) mild hypotension; (C) severe hypotension. (D) The three graphs are plotted on the same axes. (A) The relationship between PaCO2 and CBF is sigmoidal. (B) The encroachment on the vasodilatory reserve by perfusion pressure autoregulatory-induced vasodilation reduces the reactivity to carbon dioxide (CO2), as indicated by the slopes of the linear part of the sigmoid curves. Perfusion pressure autoregulation and CO2 reactivity compete to alter vessel diameter. (C) When autoregulation is exhausted, there is no further response to hypercapnia and little or no constriction to hypocapnia. (D) The midpoints of the sigmoidal curves are approximately at the resting PCO2. (E) Percent of maximal dilation. At high PCO2, or low blood pressure, vessels are maximally dilated. Healthy vessels in normotensive subjects maintain vascular tone with…

• 

  Figure 4. Analysis of ramp hypercapnia data. Stimulus paradigms in an 18-year-old male subject with a history of moyamoya disease affecting the right middle cerebral artery and both anterior cerebral artery territories. The end-tidal PCO2 (PETCO2) was initiated at baseline PETCO2, reduced to 30 mmHg by asking the subject to hyperventilate, and then raised at a constant rate, regardless of minute ventilation or breathing pattern, resulting in a linear rise of PETCO2 vs. time, to 55 mmHg over 4 min. We examined the blood oxygen-level dependent (BOLD) signal as a function of PCO2. The interrogated voxel locations are indicated by the circles on the CVR maps. Types of response: A, positive response with a sigmoidal shape; B, initial positive response which then declines; C, response which progressively declines; and D, response that initially declines but then increases. In healthy people, A Type voxels overwhelmingly predominate, and the map is substantially red. In this patient, voxels are color coded according to the color of the graph corresponding to the Type. Note the projection of CVR as would have been calculated for a 2-point PETCO2 stimulus of 45 and 50 mmHg for the voxels of Type B…

• 

  Figure 5. Calculation of voxel resistance sigmoidal profile over range of PETCO2. (A) Theoretical illustration of two brain vascular territories both supplied in parallel by a major artery with resistance (flow limitation). Their vasodilatory responses are shown as shading. (B) Simple resistance circuit model. Reference and examined vascular beds with resistances R1 and R2 are perfused via an arterial flow resistance (Rart) from mean arterial blood pressure (MAP). The pressure perfusing the two branches (Pbranch) and their respective resistances establishes flows through each branch (F1 and F2) that sum to Ftotal. Given F1 and F2 as BOLD measures for the reference and examined voxels, R1 and R2 can be calculated by considering the Model as an analog of an electrical circuit. The Model assumes that this unit may be expanded and contracted in a fractal pattern to contain the blood supply of both hemispheres, from extracranial vessels to capillaries. Every vessel or vascular bed is considered in fluid continuity with every other vessel. As such, we use the examined bed for each unit on a voxel scale, as if it is perfused in parallel with an ideal voxel of full vasodilatory responsiveness.…

• 

  Figure 6. Resistance sigmoid measures (top) and their corresponding anatomical maps (bottom). The vascular model is that described in Figure 5. Various aspects of the resistance-PETCO2 relationship are expressed in the sigmoidal shape of the relationship. The graph shows some of these parameters diagrammatically, and the maps show their anatomical distribution. The amplitude shows the full extent of the vasodilation and vasoconstriction. The midpoint of the sigmoid is the PETCO2 at its steepest slope, and the range is the span of PETCO2 over which the response can be considered linear. The reserve and sensitivity show the vasodilatory reserve available from the subject’s resting baseline PETCO2 and the response sensitivity (sigmoid slope) at resting PETCO2. The quality map shows the r2 fit of the examined voxel resistance sigmoid.

• 

  Figure 7. Normalizing the CVR score in terms of the normal range. An identical, repeatable stimulus is applied to a healthy cohort. CVR is calculated and mapped to standard space. Mean (M) and SD are calculated for each voxel. A patient is administered the same test. The patient CVR is assessed as z-scores. Z-scores can be color coded and mapped over the anatomical scan to show the distribution of abnormality (reprinted from Fisher et al., 2018 with permission of the publisher).

• 

  Figure 8. The authors and the generations of gas control devices. Left: DM with hand on Generation 1 end-tidal gas control device used in 1998. JF with hand on Generation 3 gas control device, by then called, RespirAct™. Top right: Generation 4 RespirAct™. Bottom right: Step-ramp paradigm implemented with Generation 4 RespirAct™.